Printer Friendly

Celebrity political endorsement effects: a perspective on the social distance of political parties.

Introduction

In both Taiwan (where the author conducted this study) and the United States, entertainment and politics have become increasingly intertwined (Jackson, 2008). Many actors, singers, and other influential celebrities promote their viewpoints on political issues, participate in political activities, and even endorse and campaign for candidates and political parties (Becker, 2013; Lee, Yen, Lin, Chen, & Wu, 2011; Wood & Herbst, 2007).

A rich literature on spokespersons and endorsers and their effects on consumer/commercial marketing and marketing communication is present (e.g., Aggarwal-Gupta & Dang, 2009; Chou, Lien, & Liang, 2011; Tsai & Jhuo, 2013). However, despite the pervasiveness of celebrity political endorsements (CPEs), the effective use of CPEs, the selection of suitable endorsers, and CPE effects on voters' responses are relatively underresearched in the political marketing/communication field (Brubaker, 2011; Henneberg & Chen, 2008; Morin, Ivory, & Tubbs, 2012). Furthermore, related studies have shown conflicting findings on CPE effects (Mishra & Mishra, 2014). Therefore, the primary purpose of this article is to further develop an understanding of CPE effects by examining how CPE (versus non-CPE) affects young voters' attitudes and voting behaviors.

This study focuses on young voters for four reasons. First, young people's adulation of celebrities makes them an attractive target for celebrity appeals (Jackson & Darrow, 2005). Second, young voters' attitudes and preferences regarding political candidates are often unstable (Burton & Netemeyer, 1992), which make them a campaign target (Chang & Hsieh, 2011). Third, although they show lower involvement and interest in politics and elections, communicating with the right message and through the right medium can increase young voters' engagement in the political process (Drummond, 2006). Fourth, encouraging young voters to vote is important because their voting behaviors can become habitual (Gerber, Green, & Green, 2003).

Of the different types of celebrity political endorsement, entertainer/idol endorsements and political figure endorsements are most prevalent and make strong impressions on voters (Pan, 2012; Wood & Herbst, 2007). Commercial marketing research has found that different types of endorsers and spokespersons generate different effects (e.g., Aggarwal-Gupta & Dang, 2009; Stafford, Stafford, & Day, 2002). Ohanian (1991) demonstrated that the importance and persuasiveness of source characteristics are affected by different context variables (e.g., the advertised products). Accordingly, this article assumes that political figures and idols represent different source characteristics (i.e., expertise and attractiveness, respectively), meaning that the endorsement effect should depend on the endorsed parties in political ads.

Recently, construal-level theory (CLT) has received a great deal of attention within the consumer psychology domain. This theory posits that psychological distance (i.e., a subjective perception of the distance between a person/perceiver and an event/object) systematically affects a person's mental construal (i.e., representation) of the event, leading to differing information preferences and decision modes (Trope, Liberman, & Wakslak, 2007). Psychological distance includes multiple dimensions such as temporal (distant future vs. near future), social (related to other people vs. related to oneself), spatial (remote vs. close location), and hypothetical (uncertain vs. certain) distance (Bornemann & Homburg, 2011; Liberman, Trope, & Wakslak, 2007). This article infers that the consistency or inconsistency between an endorser's recommended party and a voter's party preference (hereafter designated as party consistency) denotes social distance. Accordingly, this article's second purpose is to apply construal-level theory to the election context and explore whether party consistency moderates the effects of political endorsement ads with different endorser types. Party consistency affects voters' perceived social distance and therefore determines how voters construe party-related endorsement information.

In sum, this article examines the effects of CPE advertisements on young voters and explores whether the preference consistency of a recommended party moderates the effects of different celebrity endorser types based on the CLT mechanism. This study adds to the theoretical knowledge of CPE and social distance within CLT, and it provides suggestions on how political marketers can construct effective election strategies that incorporate celebrity endorsements.

Celebrity Political Endorsement

According to Henneberg and Chen (2008), two types of celebrities engage in political endorsements: (1) traditional/external celebrities (i.e., nonpolitical figures such as entertainers, athletes, media figures, and entrepreneurs) and (2) political/internal celebrities.

Scholars often predict/explain CPE effects by the source credibility model (focusing on endorsers' expertise and trustworthiness), source attractiveness model (stressing endorsers' likability, familiarity, and physical attractiveness; Jackson & Darrow, 2005), meaning transfer theory (McCracken, 1989), match-up hypothesis (Kamins, 1990), and balance theory (Heider, 1958).

Previous studies on CPE effects have produced conflicting findings. For instance, Austin, de Vord, Pinkleton, and Epstein (2008) found that celebrity-endorsed get-out-the-vote campaigns help to persuade young voters to participate in an election campaign by decreasing voters' complacency and increasing their self-efficacy. Henneberg and Chen's (2008) in-depth interviews revealed that most campaign managers and candidates believe that the more intense the use of CPE, the higher the possibility of winning the election. Jackson (2008) and Jackson and Darrow (2005) demonstrated that stars and athletes had favorable effects on college students' approval of endorsed political statements and positions. Nownes (2012) also found that movie stars' and athletes' implicit endorsements (i.e., financial support) affected young voters' views of the endorsed parties. Pease and Brewer (2008) and Garthwaite and Moore (2013) demonstrated the positive effects of media figure endorsements on voters' assessments of candidate viability and voting intentions and behaviors in a presidential primary. Bryce (2010) and Vining and Wilhelm (2011) found that political figure endorsements helped candidates get more votes.

In contrast, O'Regan (2014) found that college-aged adults are less likely to trust political endorsements and information from celebrities (defined as an actors, singers, talk-show hosts, and athletes) than from other sources (such as family and friends and political or interest groups), because most young adults believe celebrities are less informed about politics. Brubaker (2011) found that collegeaged voters distance themselves from celebrity endorsements of out-group candidates; however, celebrity endorsements of in-group candidates also had no influence on voters' decisions, because voters already favored the endorsed candidates. Wood and Herbst (2007) found that, compared to the influence of friends and family, idol endorsements had a negligible effect on first-time voters' voting decisions. However, Democrat supporters were in favor of idol endorsements more than Republican supporters. Payne, Hanlon, and Twomey (2007) indicated that, even though Hollywood star endorsements increased the voting intentions of young voters, they also increased the possibility of voters choosing the opponent rather than the ad sponsor. Frizzell (2011) demonstrated that Republicans and Independents were more likely to support a specific policy position advocated by a generic government official, who the experimenters set as a State Department official (i.e., a noncelebrity) without any partisan leanings, than by a singer; however, Democrats' responses to the noncelebrity versus celebrity were the same. Frizzell suggested the results might be a function of party identification, because the respondents may have assumed the State Department official was a Republican who held a similar ideology; thus, Republicans were more likely to adopt the position of this noncelebrity. However, Frizzell did not explain why the celebrity endorsement was not more effective for Democrats than the ideologically dissimilar noncelebrity.

Some possible moderators or determinants of CPE effects have been proposed to explain the conflicting findings; they include voters' personal political salience (Veer, Becirovic, & Martin, 2010), gender, ethnicity (O'Regan, 2014), perceived importance of the issues that celebrities have promoted (Becker, 2013), and endorser credibility, expertise, trustworthiness, and attractiveness (Mishra & Mishra, 2014).

Previous relevant studies have mainly examined celebrities as a generalization or have focused on only one specific type of celebrity. Few studies have compared the effects of different celebrity types on CPE. Additionally, Wood and Herbst (2007) and Frizzell (2011) found that voters' party identification seems to have an impact on celebrity political endorsement; however, the moderating effects and mechanism of voters' partisanship on CPE have not been explored. Therefore, this study investigates whether the inconsistent effects of CPE discussed in the literature are due to specific celebrity types (political figures vs. idols) and the moderating effects of party consistency.

Voters' Partisanship and Responses Toward Political Advertising

Partisanship plays a determinant role in shaping voters' responses toward political advertising and voting preference (Chang, 2003). Voters' party identification acts as perceptual screen through which advertising messages are selected and evaluated (Merritt, 1984). Voters consider political advertisements from their preferred or identified party more believable (O'Cass, 2005; Winneg, Kenski, & Jamieson, 2005). Franz and Ridout's (2007) partisan hypothesis also predicts that persuasive effects of political advertising differ depending on voters' partisanship.

Selective/biased processing and concept-similar motivated information processing are usually the functional mechanisms through which voters' party preference/identification affects the formation of advertising effectiveness responses (e.g., Chang, 2003; Chou & Chou, 2012; Stevens, Sullivan, Allen, & Alger, 2008). To maintain cognitive consistency, voters evaluate political ads sponsored by their preferred or identified party candidates as more favorable and persuasive, which increases voting support for those candidates. Nelson and Garst (2005) suggested that the consistency between the speaker's party and the recipient's party identification might evoke the recipient's central/systematic processing or function as a peripheral/heuristic cue.

In contrast to previous studies, the current research infers that the consistency or inconsistency between an endorser's recommended party and a voter's party preference also denotes social distance between the voter and the party and therefore determines how voters construe party-related endorsement information and form related attitudinal responses. Accordingly, this article uses construallevel theory to predict the possible effects of voters' partisanship.

Construal-Level Theory

An Outline of CLT and Social Distance

Construal-level theory posits that people mentally construct an event in different ways depending on the perceived psychological distance from the event (Trope & Liberman, 2003). When the event is psychologically distant, people adopt high-level construals--they think of and represent the event in easy, abstract, and decontextualized terms. Additionally, people pay more attention to the values, ends, superordinate goals, and primary features of the event. Conversely, when the event is psychologically near, people adopt low-level construals--they think of and represent the event in complex, concrete, and contextualized terms; they also pay more attention to the costs, means, subordinate goals, and secondary features of the event (Kardes, Cronley, & Kim, 2006; Trope & Liberman, 2003; Trope et al., 2007).

Differences in people's construals also affect how they evaluate events and form related predictions and judgments. As psychological distance increases, high-level construals have a greater influence on evaluations and judgments of an event than low-level construals (Nan, 2007).

One dimension of psychological distance is social distance, which is the subjective distance between a perceiver and other individuals or groups (Trope et al., 2007). Social distance takes several forms and can be manipulated in different ways. Interpersonal dissimilarity (Liviatan, Trope, & Liberman, 2008), out-group (vs. in-group) members (Trope et al., 2007; Zhao & Xie, 2011), making decisions for others rather than oneself (Bornemann & Homburg, 2011; Kim, Zhang, & Li, 2008), and increased interpersonal politeness (Stephan, Liberman, & Trope, 2010) all increase people's perceived social distance and evoke more abstract and high-level thinking and construals. For example, how an individual construes someone else's action depends on perceived psychological distance. When a stranger (distant psychological distance) gives a talk, listeners mentally represent the action in terms of abstract, superordinate goals (e.g., communicating research ideas). If a close friend (near psychological distance) gives the same talk, listeners mentally represent it in terms of subordinate, concrete means (e.g., presenting PowerPoint slides) (see Liviatan et al., 2008).

The Relationship Between People's Construing Modes and the Construed Level of Persuasive Messages

An event-related message can be construed at different levels. A high-level (low-level) construed message represents or contains more primary and core features of an event, and a low-level construed message represents or contains more secondary and surface features of an event (Kardes et al., 2006; Trope & Liberman, 2003). In general, when the format or representation of the message is consistent with the receiver's processing mode, information processing is facilitated; this also makes the message more persuasive (Schwarz & Clore, 1983). Kim, Rao, and Lee (2009) demonstrated that receivers' modes of mentally construing an event are affected by the event's temporal distance. The match between receivers' construing modes and the construed level of the event-related message improves receivers' evaluations of the message. Chou and Lien (2010); Martin, Gnoth, and Strong (2009); and Spassova and Lee (2013) redemonstrated these matching effects in different temporal distance contexts using various manipulations.

Trope and Liberman (2003) suggested that the principles applied to temporal distance could also be applied to other distance dimensions. As such, the current research described in this article assumes that similar fitness relationships exist in social distance contexts. Specifically, I propose that, when voters are exposed to political endorsement ads, a match or mismatch between voters' construing modes (influenced by the perceived social distance from the recommended party) and the construed level of the political endorsement messages (influenced by endorser types) should determine the advertising effects. Furthermore, voters' perceived social distance depends on the consistency or inconsistency between the ad-recommended party and voters' individual party preferences.

Research Hypotheses

The Effects of Party Consistency on Voters' Perceived Social Distance and Construing Mode of Party-Related Information

Interpersonal similarity occurs when people share similar attitudes, thoughts, personality characteristics, and backgrounds. It promotes a sense of belonging and closeness and thereby decreases social distance (Liviatan et al., 2008; Miller, Downs, & Prentice, 1998). The consistency (versus inconsistency) between a target political party (e.g., the party recommended in the endorsement ad) and a voter's political party preference usually represents greater similarity in mutual opinions and positions, thereby reducing social distance between the party and the perceiver. Moreover, if a voter is a supporter or member of a target party, the voter is likely to view the party and its candidates as in-group members and feel socially closer (Trope et al., 2007). According to CLT (Trope & Liberman, 2003), a voter should prefer to construe party-related information from the preference-consistent party (versus the preference-inconsistent party) in lower-level construals because of the nearer social distance. Based on these notions, the author posits the following hypothesis:

H1: The consistency (versus inconsistency) between the target party and voters' party preferences (a) decreases voters' perceived social distance from the target party and (b) makes voters construe party-related information in lower-level construals.

The Effects of Celebrity Political Endorsement

Political figures usually have more political expertise and experience (Veer et al., 2010). Endorser expertise is an important determinant of political endorsement effects (Henneberg & Chen, 2008). However, idols are perceived as more attractive, which can be transferred to the idol's endorsed candidate and party (see Wood & Herbst, 2007). The oft-noticed tendency of young people to act like their favorite celebrities suggests that idol endorsements may influence young people's political choices and responses (Jackson & Darrow, 2005; Yue & Cheung, 2000).

Based on the source credibility and attractiveness models, CPE, using either political figures or idols, should generate better endorsement effects than non-CPE (i.e., regular citizen endorsements), because regular citizens usually have comparatively less political expertise and attractiveness than political figures and idols. Additionally, young voters usually lack experience and knowledge regarding political processes and issues, which makes them susceptible to personal and opinion leadership influences (Burton & Netemeyer, 1992). Because celebrity endorsers are more viewed more favorably than noncelebrity endorsers (Veer et al., 2010), and celebrity endorsements are seen as an effective referential index for decisions (Kamins, 1990), CPE should be especially effective on younger voters. Thus, I propose the following hypothesis:

H2: In political advertising, celebrity political endorsement using political figures or idols is more effective on young voters than non-CPE (i.e., regular citizens).

The Moderation of Party Consistency on the Effects of CPE

In an election context, voters who want to select a competent candidate or party should value political expertise over attractiveness (Yu, 2003), because attractiveness is less related to the nature of political service (Henneberg & Chen, 2008). Therefore, political expertise is a high-level/primary attribute of candidates/parties, and attractiveness is a low-level/secondary attribute of candidates/parties (Kardes et al., 2006; Trope & Liberman, 2003). According to McCracken's (1989) meaning-transfer theory, as different meanings and characteristics of endorsers (i.e., expertise or attractiveness) are transferred to the endorsed targets, political figure endorsements are considered high-level construed messages, and idol endorsements are considered low-level construed messages.

When the recommended party is inconsistent with the voters' preference, the increased social distance causes voters to construe party-relevant information in high-level construals. According to Kim et al.'s (2009) match relationship, if the ad is endorsed by a political figure (i.e., a high-level construed message), voters' construing mode and the construed level of the message are matched. In contrast, if the ad is endorsed by an idol (i.e., a low-level construed message), there is a mismatch. When processing a matched message, people are more likely to experience a feeling of fluency and thus generate more positive evaluations of the message (Kim et al., 2009). Additionally, a matched message has higher perceived self-relevance and usefulness in making a decision or judgment (Martin et al., 2009; Zhao & Xie, 2011), which enhances the persuasiveness of the message. Therefore, when the recommended party is preference-inconsistent, matched political figure endorsements should generate better effects than mismatched idol endorsements. However, if the recommended party is preference-consistent, the decreased social distance causes voters to represent or construe party-relevant information in low-level construals. Idol endorsements that match voters' construing mode should be more effective than mismatched political figure endorsements. The following hypotheses describe these relationships:

H3: When the recommended party in political advertising is inconsistent with voters' party preferences, political figure endorsements are more effective than idol endorsements.

H4: When the recommended party in political advertising is consistent with voters' party preferences, idol endorsements are more effective than political figure endorsements.

Pilot Study

Because the impact of party consistency on voters' perceived social distance was the premise to infer the moderation of party consistency on CPE effects, a pilot study was conducted with a convenience sample of 103 voters who ranged in age from 20 to 30 years (the minimum voting age in Taiwan is 20) from Taipei City and New Taipei City to test H1. Following Chou and Chou (2012), a between-subjects factorial design was adopted; the independent variable was the target party (Nationalist Party [KMT] vs. Democratic Progressive Party [DPP]) in the experimental context. The KMT and the DPP are the two major political parties in Taiwan. In the 2012 elections, the KMT won 56.6% and the DPP 35.4% of the 113 seats in the Legislative Yuan (national parliament). The KMT's ideology and political position are relatively conservative and center-right; the DPP's position is more liberal and center-left.

Participants indicated their party preferences, which were used to create a new independent variable (i.e., party consistency), which compared participants' exposure to a target party and their individual party preferences. To simplify the analysis, participants who indicated a preference for no party or other (minor) parties were eliminated, resulting in a valid sample of 52 participants (51.9% men, 48.1% women) with an average age of 23.75 (SD = 2.84).

In the experimental task, participants were asked to imagine that a KMT or DPP candidate had engaged in 10 activities during an election. For each activity, participants chose one of the two restatements that best described it (i.e., testing voters' construing mode of the target party's related information); the two restatements with similar word counts were selected through a pretest (N = 48) to describe the activity in low-level or high-level construals (Liviatan et al., 2008). Next, using seven-point scales, participants assessed perceived social distance from the target party by party familiarity (ranging from not at all familiar to very familiar), party similarity ("I feel that the KMT/DPP has similar positions to my own," and "The KMT/DPP and I belong to the same group"; ranging from strongly disagree to strongly agree; a = .84), and party closeness (ranging from psychologically remote to psychologically close) (Kim et al., 2008; Stephan, Liberman, & Trope, 2011). Participants also answered demographic information questions, and, finally, they chose which party they preferred (Chang, 2003). This arrangement avoided reminding the participants of the importance of party preference, and it decreased the possibility of participants' guessing the research purpose and providing biased answers.

The analytical results revealed that party consistency (versus inconsistency) increased party familiarity (M = 4.56 vs. 3.65, t(50) = 3.13, p < .05), similarity (M = 4.45 vs. 2.68, t(50) = 8.01, p < .05), and psychological closeness (M = 3.56 vs. 1.75, t(50) = 6.03, p < .05), thereby reducing the participants' perceived social distance. These results support H1(a). Additionally, more participants in the party-consistency (versus party-inconsistency) condition interpreted the activities of the target party in low-level construals (M = .42 vs. 0.18, t(50) = 5.92, p < .05), supporting H1(b).

Experiment

Participants and Design

To test H2, H3, and H4, a two-by-three between-subjects factorial experiment was conducted to examine the advertising effects of CPE and the moderation of party consistency (recommended parties: KMT and DPP; endorser types: political figure, idol, and regular citizen). Data were gathered in two phases. The first phase was conducted before the voting day of an election to understand the immediate effects of political endorsement ads on participants' attitudinal responses. The second phase was conducted after the election to understand the ad effects on the same participants' real voting behaviors.

The recruitment and elimination criteria and methods of the sample were the same as those employed in the pilot study. Two hundred ninety-seven voters participated in the first phase of the experiment; the valid sample consisted of 137 participants (58.4% men, 41.6% women) with a mean age of 22.96 (SD = 3.26). Among these, 33 left no or illegible contact information; therefore, the second phase tracked the voting behaviors of only the remaining 104 participants and received responses from 90 participants. The responding sample (n = 90) and the drop-out sample in the second phase (n = 47) had similar demographic characteristics and no significant differences in terms of their responses to advertising communication effects in the first phase (all |t(135)| < 1.43, all p > .1).

Experimental Situation, Stimuli, and Pretest

The experiment took place during a real legislative election context (held on January 14, 2012 in Taiwan). The experimental stimulus was a printed political ad with one endorser recommending the KMT or DPP as the best choice for the nationwide constituency.

The representative male endorsers were selected based on the results of a pretest. Sixty young voters assessed pictures of these figures, answering questions on a seven-point scale. Participants assessed each figure's awareness (from low to high awareness), attractiveness (from unattractive to attractive), political expertise (from without political expertise to with political expertise), and the perceived party position of each figure (from KMT-leaning to DPP-leaning or neutral).

The representative political figure was Huai-Nan Peng, who was the president of the Central Bank of Taiwan, a governmental body that is charged with tasks such as maintaining financial stability, monetary management, foreign exchange management, and bank supervision. He was considered a possible choice for vice presidential candidate by both the KMT and the DPP in the 2012 election; this story was extensively discussed in the press. The representative idol was Zheng-Long Lan, who had starred in many dramas, movies, and advertisements. Both Huai-Nan Peng and Zheng-Long Lan had higher awareness ratings than the regular citizen (Zhi-Gi Li, a fictitious figure) (M = 5.73 = 6.07 > 1.48, F(2, 86) = 94.16, p < .05), confirming their celebrity status. The political figure had the highest political expertise ([M.sub.political figure] = 4.63 vs. [M.sub.idol] = 2.07 vs. [M.sub.citizen] = 2.79, F(2, 86) = 26.28, p < .05), and the idol had the highest attractiveness ([M.sub.idol] = 5.07 vs. [M.sub.political figure] = 3.93 vs. [M.sub.citizen] = 3.41, F(2, 86) = 7.04, p < .05) among the selected endorsers. Additionally, more than 60% of the participants thought that these three people had neutral images, suggesting that political endorsements for both political parties were plausible. Another pretest (N = 57) showed Peng's and Lan's likability were not significantly different, but both were greater than Li's ([M.sub.political figure] = 5.33 = [M.sub.idol] = 5.00 > [M.sub.citizen] = 4.13, F(2, 54) = 4.48, p < .05).

Procedure

In the first phase of the experiment--conducted from December 28, 2011, to January 7, 2012--each participant read a political endorsement ad and then completed a questionnaire that evaluated attitudinal responses. A question asking about the participant's party preference was placed at the end of the questionnaire, which was the same arrangement that had been used in previous studies (e.g., Chang, 2003; Chou & Chou, 2012). To obtain contact information for tracking voting behaviors, participants were informed at the end of the questionnaire that there would be a drawing for 100 coupons worth NT$100 each (about US$3.30), and that contact information was needed to notify the winners.

Immediately after the election, the second phase was conducted over a three-week period via email and telephone calls. After congratulating the participants for winning a coupon (every participant contacted in the second phase won a coupon) and assuring them of total anonymity, research assistants inquired about the participants' real voting behaviors and then debriefed participants, explaining the whole experiment.

First Phase Measures

Voters' attitudinal responses were measured with five advertising communication effectiveness indexes on seven-point scales, including attitudes toward the ad (from bad to good, unfavorable to favorable, and negative to positive; a = .88; Pascal, Sprott, & Muehling, 2002); the endorser (from not likable to likable,and unpleasant to pleasant; a = .92; Reinhard, Messner, & Sporer, 2006); the advertised party (from bad to good, dislike to like, and unpleasant to pleasant; a = .98; Veer et al., 2010); turnout intentions (from very unlikely to very likely and low possibility to high possibility to vote in the upcoming election; a = .98; Garramone, Atkin, Pinkleton, & Cole, 1990); and intentions to vote for the advertised party ("After viewing this ad, I will vote for the party recommended by the endorser"; ranging from strongly disagree to strongly agree; and "If the election were held tomorrow, the likelihood of me voting for the advertised party is ..."; ranging from very unlikely to very likely; a = .91; Veer et al., 2010; Pease & Brewer, 2008). As in the pilot study, participants' perceived social distance was also measured by party familiarity, similarity (a = .94), and closeness.

For manipulation checks, participants noted which party was recommended in the ad and assessed endorser awareness. Additionally, participants chose the main characteristic of the ad endorser (expertise or attractiveness or similarity) to ensure that different endorser types represented different source characteristics. Finally, to test whether political endorsements by political figures and idols belong to different construal levels, participants assessed the political information provided by the endorser using a seven-point scale adapted from Kim et al. (2009). Possible responses ranged from irrelevant to judging the (dis)advantage of the party to relevant to judging the (dis)advantage of the party, from irrelevant to the core values of the election to relevant to the core values of the election, and from irrelevant to the ultimate goals of the election to relevant to the ultimate goals of the election (a = .90).

This study measured and controlled for participants' enduring political involvement ("I do not want to miss any news about politics" and "I like to talk about politics"; a = .80; Faber, Tims, & Schmitt, 1993) and perceived political endorsement motives of the endorser, as they might affect participants' political decisions and responses to political endorsement ads (Faber et al., 1993; Jackson, 2008). Participants evaluated the following statements adapted from Yoon, Gurhan-Canli, and Schwarz (2006) and Dong, Powpaka, and Yan (2010) on a seven-point scale (ranging from extremely unlikely to extremely likely): The political endorsement by this endorser was made "as a self-marketing strategy," "for his own personal benefit," "to enhance his own self-image" (egoism motive; a = .82), "because he sincerely cares about political development," and "because of his belief that the recommended party is a better choice" (altruism motive; a = .87).

Second Phase Measures

The research assistants asked participants whether they had voted in the January 14 election, and those participants who had voted were then asked to state which party they had voted for. If they voted for the same party as the advertised party in the experiment, the answer was coded as 1; otherwise, the answer was coded as 0.

Results

Manipulation Checks. All participants correctly recognized the recommended party. The political figure and the idol were rated as having higher awareness than the regular citizen (M = 5.33, 5.65, and 1.94; F(2, 134) = 99.08, p < .05). Further, 87.0% of the participants believed the main characteristic of the political figure was expertise, and 97.7% of the participants believed the main characteristic of the idol was attractiveness. As for the regular citizen, most participants (60.4%) noted similarity ([chi square](4) = 148.65, p < .05). The participants felt that political messages provided by political figures (M = 4.55), regular citizens (M = 3.35), and idols (M = 2.62) belonged to different construal levels (F(2, 134) = 19.64, p < .05).

Hypothesis Testing. To test the hypothesis, the independent variable was set as party consistency, which was the same as in the pilot study. The analytical results show that party consistency (versus inconsistency) increased party familiarity (M = 5.69 vs. 5.02, t(135) = 3.20, p < .05), similarity (M = 4.56 vs. 2.25, t(135) = 10.96, p < .05), and closeness (M = 3.72 vs. 1.95, t(135) = 7.80, p < .05), thereby reducing the participants' perceived social distance. These results support H1(a).

Table 1 displays the results of a series of analyses of covariance (ANCOVAs) that examined the advertising effects of party consistency and endorser types. Control variables are omitted.

Party consistency positively affected participants' ad attitudes (M = 4.24 vs. 3.55), party attitudes (M = 4.67 vs. 2.81), and intentions to vote for the advertised party (M = 4.71 vs. 2.28) (all F(1, 128) > 9.68, all p < .05). However, it did not affect endorser attitudes or turnout intentions (both F(1, 128) < 0.54, both p > .1).

Endorser types significantly affected ad attitudes, endorser attitudes, party attitudes, and turnout intentions (all F(2, 128) > 3.16, all p < .05). Political figure endorsements and idol endorsements, in contrast to regular citizen endorsements, increased participants' ad attitudes (M = 4.19, 4.17 > 3.42), endorser attitudes (M = 4.32, 4.77 > 3.21), party attitudes (M = 4.10, 4.12 > 3.19), and turnout intentions (M = 6.41, 6.37 > 5.78) (all p < .05). That is, celebrity political endorsement by political figures and idols was more effective than non-CPE, supporting H2. Additionally, participants had more favorable endorser attitudes toward idols than political figures (p < .05). The results of planned contrast indicated that idol endorsements, as compared to regular citizen endorsements, increased participants' intentions of voting for the advertised party (M = 3.95 vs. 3.03, p = .055 < .1).

More importantly, party consistency and endorser types had significant interactive effects on ad attitudes, endorser attitudes, and party attitudes (as shown in Figure 1), and intentions to vote for the advertised party (as shown in Figure 2) (all F(2, 128) > 3.60, all p < .05). As shown in Table 2, under the party-inconsistency condition, political figure endorsements generated better effects on ad attitudes (M = 4.27 vs. 3.46, p < .05), party attitudes (M = 3.60 vs. 2.60, p < .05), and voting intentions toward the advertised party (M = 2.78 vs. 2.03; p = .085 < .1) than did idol endorsements, supporting H3. Regular citizen endorsements evoked worse ad attitudes, endorser attitudes, and party attitudes than did political figure endorsements (all p < .05), but generated similar effects to idol endorsements on most advertising effectiveness indices (all p > .1) aside from endorser attitudes (M = 2.83 vs. 4.53, p < .05).

Under the party-consistency condition, idol endorsements, compared to political figure endorsements, generated better effects on improving ad attitudes (M = 4.74 vs. 4.13), endorser attitudes (M = 4.96 vs. 4.08), party attitudes (M = 5.32 vs. 4.49), and intentions to vote for the advertised party (M = 5.48 vs. 4.46) (all p < .05), supporting H4. Regular citizen endorsements evoked worse ad attitudes, endorser attitudes, and party attitudes, as well as lower voting intentions toward the advertised party than idol endorsements (all p < .05), but generated similar effects to political figure endorsements on all advertising effectiveness indexes (all p > .1).

Effects of the Experimental Variables on Voting Behaviors. Analytical results of binary logistical regressions show that party consistency increased the probability that participants voted for the advertised party ([beta] = 2.15, Wald [chi square] (1) = 13.76, p < .05, Exp(B) = 8.62). Additionally, compared to regular citizen endorsements, both idol ([beta] = 1.25, Wald [chi square] (1) = 3.05, p = .081 < .1, Exp(B) = 3.48) and political figure endorsements (P = 0.43, Wald [chi square] (1) = 0.46, p > .1, Exp(B) = 1.54) increased the likelihood that participants would vote for the advertised party; however, the effects for the latter were not significant.

Under the party-inconsistency condition, when the endorser was a regular citizen ([beta] = -2.99, Wald [chi square] (1) = 3.70, p = .055 < .1, Exp(B) = 0.05) or an idol ([beta] = -1.30, Wald [chi square] (1) = 0.71, p > .1, Exp(B) = 0.27) rather than a political figure, the probability of voting for the advertised party decreased. For political figure endorsements, this probability was 33.3%; for idol endorsements, 20%; and for regular citizen endorsements, 10%. Under the party-consistency condition, when the endorser was a regular citizen ([beta] = -2.16, Wald [chi square] (1) = 2.84, p = .092 < .1, Exp(B) = 0.11) or a political figure ([beta] = -2.56, Wald [chi square] (1) = 3.85, p < .05, Exp(B) = 0.08) rather than an idol, the probability of voting for the advertised party decreased significantly: from 88.9% for idol endorsements to 58.3% and 55% for regular citizen and political figure endorsements, respectively.

General Discussion

Conclusions

This study examined the advertising effects of celebrity political endorsement (versus non-CPE) on young Taiwanese voters and explored whether party consistency moderated the effects of different celebrity endorser types. Several conclusions can be drawn from the experimental results. First, political messages delivered by political figures, regular citizens, and idols belong to comparatively high-level, moderate-level, and low-level construals, respectively. Because of greater similarity with regular citizens, voters might believe that the preferences of regular citizen endorsers apply to themselves (O'keefe, 1990). Additionally, messages delivered by regular citizen endorsers have greater self-relevance (Bickart & Schindler, 2001) and are valuable for decision reference. Therefore, although similarity is not as relevant to political decisions as expertise, it is more relevant than attractiveness and belongs to moderate-level construals.

Second, in addition to having a direct impact on improving voters' attitudinal responses, party consistency reduces voters' perceived social distance from the target party and entices voters to view party-related information in lower-level construals. It is worth noting that party consistency does not affect young voters' endorser attitudes.

Third, political ads endorsed by political figures or idols (versus regular citizens) improve young voters' attitudes toward the ad, endorser, and party and increase turnout intentions. Additionally, idol endorsements increase intentions to vote for the advertised party more than regular citizen endorsements.

Fourth, voters respond better to ads featuring political figure endorsements that recommend a preference-inconsistent party, but they respond better to ads featuring idol endorsements in ads for a preference-consistent party.

Fifth, under the party-inconsistency (party-consistency) condition, regular citizen endorsements generate similar effects to idol (political figure) endorsements, but worse effects than political figure (idol) endorsements. We might infer from this that voters can discriminate only between match and mismatch conditions but not the match degree (e.g., high, moderate, or low). Therefore, moderate-degree matches (i.e., regular citizen endorsements) and low-degree matches (i.e., idol endorsements in party-inconsistency and political figure endorsements in party-consistency conditions) are considered mismatches and generate similarly worse effects. Additionally, psychological distance exerts an influence on the salience of low-level construals when the operationalization of the construal level is sufficiently low and on high-level construals when the operationalization of the construal level is sufficiently high (Nan, 2007). The weaker effects of regular citizen endorsements under both far and near social distance conditions may be due to their construal levels being moderate.

Finally, political endorsement ads influence young voters' real voting behaviors. Although some effects are only directionally consistent, even tiny changes are valuable for candidates, and these can alter election results when competition is fierce (Lau, Sigelman, Heldman, & Babbitt, 1999).

Theoretical Implications

This study has several theoretical implications. First, it expands and deepens the political marketing/communication literature by examining the application and causal effects of celebrity political endorsements outside U.S. and European political contexts, and it adopts multiple advertising effectiveness indexes, which responds to Pease and Brewer's (2008) suggestion that CPE effects should be examined comprehensively. Additionally, most studies on CPE are cross-sectional (e.g., Mishra & Mishra, 2014; Morin et al., 2012; Pease & Brewer, 2008; Veer et al., 2010); therefore, the two-phase design of this study demonstrates longitudinally that CPE has substantial effects on young voters' actual voting behaviors. The results affirm the importance of this research subject and provide practical value for election campaigns.

Second, this article demonstrates that celebrity type is an important determinant of political endorsement effects, but the consistency or inconsistency between the advertised party and voters' party preference moderates the endorsed effects. These findings help clarify and offer possible explanations for the mixed CPE effects found in previous studies (e.g., Brubaker, 2011; Jackson, 2008; Payne et al., 2007; Vining & Wilhelm, 2011). The current research is also the first to empirically demonstrate the moderation of party consistency on relative endorsement effects of different celebrity types. Furthermore, although the impact of party preference and identification on the formation of voters' ad responses have been documented in traditional political advertising research (e.g., Chang, 2003; Chou & Chou, 2012; Franz & Ridout, 2007), the functional mechanism is limited to selective/biased processing. The current research is based on a completely different theoretical perspective (i.e., social distance in construal-level theory). Thus, this study adds to current understanding of the role of parties on the political persuasion process and suggests that future studies should place more emphasis on the different mechanisms of party preference/identification.

Third, most CPE studies use classical product endorser theories to infer and explain political endorsement effects (e.g., Henneberg & Chen, 2008; Jackson, 2008; Morin et al., 2012; Wood & Herbst, 2007). In commercial marketing, considerable research on the relative effects of spokesperson/endorser types has focused on the simple match-up relationship in characteristics or images between spokesperson/endorser types and product/service types (e.g., Aggarwal-Gupta & Dang, 2009; Stafford et al., 2002; Tsai & Jhuo, 2013). The current article is an initial effort to apply an emerging theory, CLT, to examine the effects of endorsers. This study demonstrates the construal-level relationship among endorser types and the moderating role of perceived psychological distance. Thus, it provides new insights into endorsements in political marketing and commercial marketing domains. As such, scholars can reexamine endorser effects in a different match-up viewpoint (i.e., the match between people's construing modes and the construed level of persuasive messages).

Fourth, despite differences among existing psychological distance dimensions (Bornemann & Homburg, 2011; Liberman et al., 2007), the most widely explored dimension is still temporal distance. Additionally, few studies have explored the application of CLT in the advertising domain (Martin et al., 2009). Therefore, this article contributes to the development of the social distance concept in CLT by applying it to election and political endorsement advertising contexts. The results demonstrate that party consistency is a determinant of perceived social distance, which can further affect voters' construals regarding the target party. This adds to the understanding of possible distance antecedents.

Finally, this study expands the theoretical applicability of Kim et al.'s (2009) match/mismatch relationship to social distance contexts. It also provides preliminary evidence of the effects of moderatelevel construed messages, which was neglected in previous match/mismatch studies (e.g., Chou & Lien, 2010; Kim et al., 2009; Martin et al., 2009).

Practical Suggestions

This article provides several helpful suggestions for practitioners. First, politicians and political parties from democracies around the world use various methods to increase turnout among young voters (Drummond, 2006; Hargittai & Shaw, 2013; Leppaniemi, Karjaluoto, Lehto, & Goman, 2010). Young voters' unstable attitudes and preferences regarding political candidates mean that young voters have become a key source of votes for candidates and parties. The findings of this study reveal that young voters are suitable targets for CPE ads, and CPE improves their attitudinal responses, turnout intentions, and the probability of voting for the advertised party. This suggests that political ads that target young voters should employ celebrity endorsers. Additionally, parties and candidates should give priority to cooperating with idols followed by political figures.

Second, Mishra and Mishra (2014) suggested that celebrity endorsers should have perceived credibility so that their endorsements build political (i.e., party) brand equity. The results of the current study suggest that political marketers further consider and differentiate the dimensions of endorser credibility (i.e., attractiveness, expertise, and trustworthiness), because different types of endorsers might generate varied voter responses.

Third, political endorsement advertising generates more favorable effects on party-consistent than party-inconsistent voters. Thus, political marketers should target party-consistent voters with political endorsement advertising.

Fourth, political marketers should also consider the target audience's party preferences when selecting an endorser. Because many media outlets tend to favor a specific party, their audiences are more oriented toward that party. When most of the audience of a specific media outlet is party-consistent with the advertised candidate or party, idols make better endorsers. However, when most of the audience is party-inconsistent, known political figures who lack obvious party orientations can generate better effects than other endorser types.

Fifth, the findings indicate that the preference (in)consistency of the endorsed party does not affect young voters' endorser attitudes. Therefore, when a celebrity considers whether he or she should engage in political endorsement, thereby aligning with a specific party, concerns about displeasing young supporters of the opposing party should not be overemphasized.

Sixth, because voters' information-construing modes vary depending on the advertised party, political marketers should provide other messages in a matched construal level to generate more favorable advertising effects.

Finally, the increasing use of celebrity endorsers has generated some controversy, including claims of distracting voters from important issues and the candidates' political views and trivializing the campaign process (Pease & Brewer, 2008). However, Veer et al. (2010) have suggested that celebrities can be used to encourage voters to engage with political issues and views. The current study's results imply that selecting a more likable or appropriate endorser may increase voters' engagement with important issues.

Limitations and Future Research

This study has some limitations. First, the study focused only on young voters. Future research should test a sample of older voters to increase the generalizability of the findings. Moreover, future research could segment young voters into several subgroups (e.g., students vs. nonstudents, or rural vs. urban) to further examine CPE effects on different young voters. Tests also might be conducted on the possible moderating effects of other voter-related variables (e.g., general attitudes toward political ads, political interest, and public mood).

Second, this study focused only on voters whose preferred party was either the KMT or the DPP. Future research can examine the perceived social distance that supporters of minor parties and independent voters hold toward the two major parties.

Third, employing celebrities to transfer their symbolic value to the endorsement process might be more consistent with the communicative mode in collectivistic cultures (e.g., Asian countries) than in individualistic cultures (e.g., Western countries) (Choi, Lee, & Kim, 2005). Future research should replicate the experiments in Western countries and other Asian countries to enhance the generalizability and identify culture's possible moderating effect on CPE effects.

Finally, the political endorsement effects and the corresponding construal levels of other celebrity types (e.g., media figures, entrepreneurs, and athletes) and idol types (e.g., singers, models, and hosts) are also worthy of further exploration.

References

Aggarwal-Gupta, M., & Dang, P. J. (2009). Examining celebrity expertise and advertising effectiveness in India. South Asian Journal of Management, 16(2), 61-75.

Austin, E. W., de Vord, R. V., Pinkleton, B. E., & Epstein, E. (2008). Celebrity endorsements and their potential to motivate young voters. Mass Communication and Society, 11, 420-436.

Becker, A. B. (2013). Star power? Advocacy, receptivity, and viewpoints on celebrity involvement in issue politics. Atlantic Journal of Communication, 21(1), 1-16.

Bickart, B., & Schindler, R. M. (2001). Internet forums as influential sources of consumer information. Journal of Interactive Marketing, 15(3), 31-40.

Bornemann, T., & Homburg, C. (2011). Psychological distance and the dual role of price. Journal of Consumer Research, 38, 490-504.

Brubaker, J. (2011). It doesn't affect my vote: Third-person effects of celebrity endorsements on college voters in the 2004 and 2008 presidential elections. American Communication Journal, 13(2), 422.

Bryce, S. (2010). The endorsement effect: An examination of statewide political endorsements in the 2008 democratic caucus and primary season. American Behavioral Scientist, 54, 284-297.

Burton, S., & Netemeyer, R. G. (1992). The effect of enduring, situational, and response involvement on preference stability in the context of voting behavior. Psychology & Marketing, 9, 143-157.

Chang, C. C. (2003). Party bias in political advertising processing: Results from an experiment involving the 1998 Taipei mayoral election. Journal of Advertising, 32(2), 55-67.

Chang, C. H., & Hsieh, C. W. (2011, November 24). In 2012 election, blue and green parties fight for 1.2 million first-time voters. PTS News. Retrieved from http://ma19.cc/2011042500005

Choi, S. M., Lee, W. N., & Kim, H. J. (2005). Lessons from the rich and famous: A cross-cultural comparison of celebrity endorsement in advertising. Journal of Advertising, 34(2), 85-98.

Chou, H. Y., & Chou, J. S. (2012). The advertising effectiveness of political text messages: An exploratory study on undergraduates from five universities in north Taiwan. Taiwanese Journal of Political Science, 54, 71-120.

Chou, H. Y., & Lien, N. H. (2010). The impact of temporal distance on the effects of negative campaign advertisements. Taiwan Democracy Quarterly, 7(2), 33-76.

Chou, H.-Y., Lien, N.-H., & Liang, K.-Y. (2011). The antecedents and belief-polarized effects of thought confidence. Journal of Psychology, 145, 481-506.

Dong, L., Powpaka, S., & Yan, K. (2010). Corporate social responsibility based on consumers' viewpoints: Altruism or self-interest. Chinese Journal of Management, 7, 861-867.

Drummond, G. (2006). Political parties' use of web based marketing: Some preliminary findings relating to first-time voters in the 2005 general election. International Journal of Nonprofit and Voluntary Sector Marketing, 11, 181-191.

Faber, R. J., Tims, A. R., & Schmitt, K. G. (1993). Negative political advertising and voting intent: The role of involvement and alternative information sources. Journal of Advertising, 22(4), 67-76.

Franz, M. M., & Ridout, T. N. (2007). Does political advertising persuade? Political Behavior, 29, 465-491.

Frizzell, C. (2011). Public opinion and foreign policy: The effect of celebrity endorsements. Social Science Journal, 48, 314-323.

Garramone, G. M., Atkin, C. K., Pinkleton, B., & Cole, R. T. (1990). Effects of negative political advertising on the political process. Journal of Broadcasting & Electronic Media, 34(3), 299-311.

Garthwaite, C., & Moore, T. J. (2013). Can celebrity endorsements affect political outcomes? Evidence from the 2008 US democratic presidential primary. Journal of Law, Economics & Organization, 29, 355-384.

Gerber, A. S., Green, D. P., & Green, M. (2003). Partisan mail and voter turnout: Results from randomized filed experiments. Electoral Studies, 22, 563-579.

Hargittai, E., & Shaw, A. (2013). Digital savvy citizenship: The role of Internet skills and engagement in young adults' political participation around the 2008 presidential election. Journal of Broadcasting & Electronic Media, 57, 115-134.

Heider, F. (1958). The psychology of interpersonal relations. New York, NY: John Wiley.

Henneberg, S. C., & Chen, Y. L. (2008). Celebrity political endorsement: Campaign management for the Taipei City councillor election 2002. Journal of Political Marketing, 6(4), 1-31.

Jackson, D. J. (2008). Selling politics: The impact of celebrities' political beliefs on young Americans. Journal of Political Marketing, 6(4), 67-83.

Jackson, D. J., & Darrow, T. I. A. (2005). The influence of celebrity endorsements on young adults' political opinions. Harvard International Journal of Press/Politics, 10(3), 80-98.

Kamins, M. A. (1990). An investigation into the "match-up" hypothesis in celebrity advertising: When beauty may be only skin deep. Journal of Advertising, 19(1), 4-13.

Kardes, F. R., Cronley, M. L., & Kim, J. (2006). Construal-level effects on preference stability, preferencebehavior correspondence, and the suppression of competing brands. Journal of Consumer Psychology, 16, 135-144.

Kim, H., Rao, A. R., & Lee, A. Y. (2009). It's time to vote: The effect of matching message orientation and temporal frame on political persuasion. Journal of Consumer Research, 35, 877-889.

Kim, K., Zhang, M., & Li, X. (2008). Effects of temporal and social distance on consumer evaluations. Journal of Consumer Research, 35, 706-713.

Lau, R. R., Sigelman, L., Heldman, C., & Babbitt, P. (1999). The effects of negative political advertisements: A meta-analytic assessment. American Political Science Review, 93, 851-875.

Lee, C. S., Yen, Y. L., Lin, C. C., Chen, Y. C., & Wu, C. C. (2011, December 20). Star endorsement increases popularity of legislator candidates. China Times. Retrieved from http://vote.chinatimes.com/2012/election/content.aspx?cat=15&art=19551

Leppaniemi, M., Karjaluoto, H., Lehto, H., & Goman, A. (2010). Targeting young voters in a political campaign: Empirical insights into an interactive digital marketing campaign in the 2007 Finnish general election. Journal of Nonprofit and Public Sector Marketing, 22, 14-37.

Liberman, N., Trope, Y., & Wakslak, C. (2007). Construal level theory and consumer behavior. Journal of Consumer Psychology, 17, 113-117.

Liviatan, I., Trope, Y., & Liberman, N. (2008). Interpersonal similarity as a social distance dimension: Implications for perception of others' actions. Journal of Experimental Social Psychology, 44, 1256-1269.

Martin, B. A. S., Gnoth, J., & Strong, C. (2009). Temporal construal in advertising: The moderating role of temporal orientation and attribute importance in consumer evaluations. Journal of Advertising, 38(3), 5-19.

McCracken, G. (1989). Who is the celebrity endorser? Cultural foundations of the endorsement process. Journal of Consumer Research, 16, 310-321.

Merritt, S. (1984). Negative political advertising: Some empirical findings. Journal of Advertising, 13(3), 27-38.

Miller, D. T., Downs, J. S., & Prentice, D. A. (1998). Minimal conditions for the creation of a unit relationship: The social bond between birthdaymates. European Journal of Social Psychology, 28, 475-481.

Mishra, A. A., & Mishra, A. (2014). National vs. local celebrity endorsement and politics. International Journal of Politics, Culture, and Society. Advance online publication. doi:10.1007/s10767-0149174-y

Morin, D. T., Ivory, J. D., & Tubbs, M. (2012). Celebrity and politics: Effects of endorser credibility and sex on voter attitudes, perceptions, and behaviors. Social Science Journal, 49, 413-420.

Nan, X. I. (2007). Social distance, framing, and judgment: A construal level perspective. Human Communication Research, 33, 489-514.

Nelson, T. E., & Garst, J. (2005). Values-based political messages and persuasion: Relationships among speaker, recipient, and evoked values. Political Psychology, 26, 489-515.

Nownes, A. J. (2012). An experimental investigation of the effects of celebrity support for political parties in the United States. American Politics Research, 40, 476-500.

Ohanian, R. (1991). The impact of celebrity spokespersons' perceived image on consumers' intention to purchase. Journal of Advertising Research, 31(1), 46-54.

O'Cass, A. (2005). Political campaign advertising: Believe it or not. Journal of Nonprofit and Public Sector Marketing, 14, 205-221.

O'keefe, D. J. (1990). Persuasion theory and research. Newbury Park, CA: SAGE Publications.

O'Regan, V. R. (2014). The celebrity influence: Do people really care what they think? Celebrity Studies. Retrieved from http://www.tandfonline.com/doi/abs/10.1080/19392397.2014.925408#.UdtX_mSyyU

Pan, C. T. (2012, January 18). Conceptual platforms: There's still a difference between scientists and politics. China Times. Retrieved from http://rbhouse.blog126.fc2.com/blog-entry-1318.html

Pascal, V. J., Sprott, D. E., & Muehling, D. D. (2002). The influence of evoked nostalgia on consumers' responses to advertising: An exploratory study. Journal of Current Issues and Research in Advertising, 24(1), 39-49.

Payne, J. G., Hanlon, J. P., & Twomey III, D. P. (2007). Celebrity spectacle influence on young voters in the 2004 presidential campaign. American Behavioral Scientist, 50, 1239-1246.

Pease, A., & Brewer, P. R. (2008). The Oprah factor: The effects of a celebrity endorsement in a presidential primary campaign. International Journal of Press/Politics, 13, 386-400.

Reinhard, M. A., Messner, M., & Sporer, S. L. (2006). Explicit persuasive intent and its impact on success at persuasion: The determining roles of attractiveness and likeableness. Journal of Consumer Psychology, 16, 249-259.

Schwarz, N., & Clore, G. L. (1983). Mood, misattribution, and judgments of well-being: Informative and directive functions of affective states. Journal of Personality and Social Psychology, 45, 513-523.

Spassova, G., & Lee, A. Y. (2013). Looking into the future: A match between self-view and temporal distance. Journal of Consumer Research, 40, 159-171.

Stafford, M. R., Stafford, T. F., & Day, E. (2002). A contingency approach: The effects of spokesperson type and service type on service advertising perceptions. Journal of Advertising, 31(2), 17-34.

Stephan, E., Liberman, N., & Trope, Y. (2010). Politeness and psychological distance: A construal level perspective. Journal of Personality and Social Psychology, 98, 268-280.

Stephan, E., Liberman, N., & Trope, Y. (2011). The effects of time perspective and level of construal on social distance. Journal of Experimental Social Psychology, 47, 397-402.

Stevens, D., Sullivan, J., Allen, B., & Alger, D. (2008). What's good for the goose is bad for the gander: Negative political advertising, partisanship, and turnout. Journal of Politics, 7, 527-541.

Trope, Y., & Liberman, N. (2003). Temporal construal. Psychological Review, 110, 403-421.

Trope, Y., Liberman, N., & Wakslak, C. (2007). Construal levels and psychological distance: Effects on representation, prediction, evaluation, and behavior. Journal of Consumer Psychology, 17, 83-95.

Tsai, C.-C., & Jhuo, J.-Y. (2013). The research on the match-up between spokesperson and product types. Journal of Management and Systems, 20, 611-635.

Veer, E., Becirovic, I., & Martin, B. A. S. (2010). If Kate voted conservative, would you? The role of celebrity endorsements in political party advertising. European Journal of Marketing, 44, 436-450.

Vining, R. L., & Wilhelm, T. (2011). The causes and consequences of gubernatorial endorsements: Evidence from state supreme court elections. American Politics Research, 39, 1072-1096.

Winneg, K., Kenski, K., & Jamieson, K. H. (2005). Detecting the effects of deceptive presidential advertisements in the spring of 2004. American Behavioral Scientist, 49, 114-129.

Wood, N. T., & Herbst, K. C. (2007). Political star power and political parties: Does celebrity endorsement win first-time votes? Journal of Political Marketing, 6(2/3), 141-158.

Yoon, Y., Gurhan-Canli, Z., & Schwarz, N. (2006). The effect of corporate social responsibility (CSR) activities on companies with bad reputations. Journal of Consumer Psychology, 16, 377-390.

Yu, C. H. (2003). Exploring the electorate's ideal candidate: The case of 2000 presidential election in Taiwan. Soochow Journal of Political Science, 17, 93-120.

Yue, X. D., & Cheung, C. K. (2000). Selection of favorite idols and models among Chinese young people: A comparative study in Hong Kong and Nanjing. International Journal of Behavioral Development, 24, 91-98.

Zhao, M., & Xie, J. (2011). Effects of social and temporal distance on consumers' responses to peer recommendations. Journal of Marketing Research, 48, 486-496.

HSUAN-YI CHOU (1)

National Sun Yat-sen University, Taiwan

Hsuan-Yi Chou: hsuanyi@mail.nsysu.edu.tw

Date submitted: 2014-03-14

(1) This research was supported by the National Science Council, Taiwan, project No. NSC 100-2410-H-110069. The author would also like to thank the anonymous reviewers for their valuable feedback.

Table 1. ANCOVA Results.

                     Party consistency        Endorser types

                     ANCOVA    Significance   ANCOVA    Significance
                     F value                  F value

Ad attitudes         9.681     .002 **        3.679     .028 *
Endorser attitudes   .536      .466           19.176    .000 **
Party attitudes      85.638    .000 **        3.814     .025 *
Turnout intentions   .103      .749           3.158     .046 *
Intentions to vote   89.794    .000 **        1.888     .155
  for the
  advertised party

                     Party consistency x
                     Endorser types

                     ANCOVA    Significance
                     F value

Ad attitudes         4.043     .020 *
Endorser attitudes   3.598     .030 *
Party attitudes      6.578     .002 **
Turnout intentions   .063      .939
Intentions to vote   4.614     .012 *
  for the
  advertised party

* p < .05. ** p < .01.

Table 2. Means for the Party-Inconsistency and Party-Consistency
Groups.

                     Party inconsistency

                     Political   Idol   Regular
                     figure             citizen

Ad attitudes         4.27        3.46   3.06
Endorser attitudes   4.63        4.53   2.83
Party attitudes      3.60        2.60   2.35
Turnout intentions   6.53        6.42   5.79
Intentions to vote   2.78        2.03   2.08
  for the
  advertised party

                     Party consistency

                     Political   Idol   Regular
                     figure             citizen

Ad attitudes         4.13        4.74   3.83
Endorser attitudes   4.08        4.96   3.66
Party attitudes      4.49        5.32   4.18
Turnout intentions   6.33        6.33   5.77
Intentions to vote   4.46        5.48   4.16
  for the
  advertised party
COPYRIGHT 2015 University of Southern California, Annenberg School for Communication & Journalism, Annenberg Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2015 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Chou, Hsuan-Yi
Publication:International journal of communication (Online)
Article Type:Report
Date:Jan 1, 2015
Words:9442
Previous Article:Violent frames: analyzing internet movie database reviewers' text descriptions of media violence and gender differences from 39 years of U.S. action,...
Next Article:Media market research on immigrant audiences: lessons learned from a critical analysis of the Spanish media survey.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters